Field test results using the BioPEPR cervical smear prescreening system

Zahniser, David J.; Oud, Peter S.; Raaijmakers, M.C.; Vooys, G. P.; Walle, R. T. Van de

doi:10.1002/cyto.990010305

Cited by 16 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, the severe dysplasia and invasive carcinoma classes had the lowest correct classification, 67% and 60%, respectively. Therefore, there was not the strong correlation between lower classification errors and severity of disease suggested in other studies (4,18,22). Indeed, there was a tendency for increasingly correct classification in the two lesser dysplasia classes, mild and moderate.…”

Section: Discussionmentioning

confidence: 63%

“…Therefore, the training set of abnormal objects probably contained a large proportion of normal objects which were globally labelled abnormal. This is reflected by the relatively high error rates obtained in the single-cells training classifiers when compared to results of the traditional CYBEST single-cell classifiers (2,16,18,19,22), where cells are labelled a priori, individually by class, during the training phase.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of contextual analysis for computer classification of cervical smears

et al. 1987

View full text Add to dashboard Cite

A procedure for automated analysis of cervical smears has been implemented in an image cytometry system. Smears are described exclusively in terms of global and contextual information extracted by pattern-recognition algorithms and represented by a vector ofproportions of cellular object types. Linear discriminant functions, based on a Fisher criterion, are derived to classify smears with a cross-section of diagnoses into two broad categories, normal and abnormal. Results obtained from 83 smears indicate 78% correct classification. In contrast to most automated systems, good classification results were obtained in normal smears with benign changes caused by inflammation and with postmenopausal atrophia and in abnormals with mild dysplasia. These findings suggest that contextual analysis may be sensitive to subtle changes in cellular morphology and to progressive patterns of dysplasia. When used with standard isolated cell analysis, contextual analysis may provide additional complementary information for automated cervical prescreening.Key terms: Automated cytology, cervical smear prescreening, image analysis, image cytometry, contextual analysis, cell classification, pattern recognition Present automated prescreening systems rely solely on the frequency of normal and abnormal cell classes present on the smear, as determined by a single-cell classifier, to establish a smear diagnosis. This method of smear prescreening often requires laborious computations that make it difficult to satisfy the basic aims in automation; speed and accuracy. Complex algorithms for excluding cell clusters and other cellular artifacts are required to guarantee successful recognition of isolated cells. In addition to the computation burden of these algorithms, they suppress cytological information from the smear classifier decision rule.Although cytotechnologists and pathologists utilize the frequency of cell classes to form a positive diagnosis (3,10), they also search for certain contextual patterns that are useful in describing the overall appearance of a cervical smear (15). Examples of contextual patterns include the abundance and spatial distribution on a smear of single cells, cell clusters, and cellular derivatives (i.e. degenerated cytoplasms and nuclei extruded from cells). Moreover, a recent study (2) suggests that human screeners achieve a lower accuracy in smear classification when visual observations of smears are made using only isolated cells as opposed to using contextual information in addition to isolated cell data. This indicates that a more global approach to smear classification using contextual information may provide automated devices with better accuracy and reduced processing time. This article examines the usefulness of contextual analysis in a computer-assisted smear presweening system. MATERIALS AND METHODS Smear Preparation and StainingThe specimens were collected from second scrapes taken immediately after the first smear from every second or third woman participating in the Nijmegen Population Screening P...

show abstract

Section: Discussionmentioning

confidence: 63%

Section: Discussionmentioning

confidence: 99%

Evaluation of contextual analysis for computer classification of cervical smears

et al. 1987

View full text Add to dashboard Cite

show abstract

“…Earlier cytophotometric studies of CIN lesions concentrated mostly on features that could discriminate between 1) normal and abnormal cell specimen, 2) CIN I1 specimen and CIN I11 or invasive carcinoma specimen, and 3) intermediate cells of "normal" cytological specimen and normal appearing intermediate cells in cytological specimen with concomitant dysplastic cells (3,4,9,11,14,15,28,29,30,31). Wheeler and co-workers were able to correctly classify cells from CIN 11, CIN 111, and invasive carcinoma lesions in about 90% of cases, using cellular and nuclear geometrical, density, and texture features (30).…”

Section: Discussionmentioning

confidence: 99%

Cytophotometric analysis of cervical intraepithelial neoplasia grade III, with and without synchronous invasive squamous cell carcinoma

et al. 1990

View full text Add to dashboard Cite

Cytophotometric analysis was performed in nuclei retrieved from paraffinembedded cervical tissue from 57 cases of CIN 111. CIN 111 lesions of patients without invasive squamous cell carcinoma (N = 37) were regarded to represent a mixture of progressive and nonprogressive lesions. The CIN 111 lesions of patients with a synchronous invasive squamous cell carcinoma (N = 20) were regarded as representing truly progressive precursor lesions (C1N.INV). Twenty-one photometric features describing geometrical, density, and texture characteristics were extracted from the digitized nuclear images. Statistical analysis of cytophotometric data indicated significant differences between the group of CIN 111 lesions and CIN.INV lesions. A cluster analysis, using one co-occurrency texture feature (S-HOMOG), one density feature (S-DI), and two geometrical features (S-AREA and M-CIRC), showed that two clusters (Cl and C2) were present in the total group of CIN 111 and CIN.INV lesions. The vast majority of CIN.INV lesions was member of one and the same cluster C1. The CIN I11 group appeared to consist of a mixture of two clusters, 54% C1 and 46% C2 lesions. Of patients 45 years or younger, the majority (62%) of CIN 111 lesions had feature values, corresponding with those of cluster Cl, and as such possibly with a potentially progressive course. In patients older than 45 years the percentage of CIN 111 lesions with C1 feature values was 27%.

show abstract

“…Due to the increased data acquisition and processing capabilities, entire cytological specimens are now ready to be analyzed. At the present time systems are operational for the screening of cervical smears (8,23,27) selection of well-spread metaphases in the field of cytogenetics (6), and automated analysis of blood smears (13,14).…”

Section: Reduction Of the Number Of Nonmitoticmentioning

confidence: 99%

Image processing for mitoses in sections of breast cancer: A feasibility study

et al. 1984

View full text Add to dashboard Cite

This paper describes an image analysis technique for the counting of nuclei in mitosis in tissue sections. Five experienced pathologists scored mitoses in photographs of preselected areas of tissue sections of the breast. Objects consistently labelled as mitotic cells by all five pathologists were considered "mitoses" in the analysis. In total, there were 45 mitotic nuclei, 68 possible mitotic nuclei and 1,172 nonmitotic nuclei.The image analysis procedure was designed to give priority to a low false negative rate, i.e., misclassification of mitoses. The procedure consists of three steps:1. Segmentation of the image. Reduction of the number of nonmitoticnuclei by using feature values based on the brightness histogram of the objects. 3. Fully automatic classification of the remaining objects using contour features.The objects remaining after the first two steps were visualized in a composite display for interactive evaluation: 10% of the mitotic nuclei were missed, and 85% of the nonmitotic nuclei were eliminated. The result of the fully automatic procedure described in this paper is rather disappointing and gave a loss of 37% of the mitoses while 5% of the nonmitotic nuclei remained.Key terms: Image analysis, mitotic activity, tissue sections, automated analysisIn the sixties, when digital image processing entered the biomedical area, single cells were among the first topics to be studied. Due to the increased data acquisition and processing capabilities, entire cytological specimens are now ready to be analyzed. At the present time systems are operational for the screening of cervical smears (8,23,27) selection of well-spread metaphases in the field of cytogenetics (6), and automated analysis of blood smears (13,14).In these screening systems, the use of single cell specimens in a monolayer suspension still is essential. Monolayers are obviously not present in a major part of diagnostic practice: tissue section analysis. As a result of this complication, the literature on the analysis of tissue sections is quite limited. The main issues have been the nuclear architecture of the epithelium of urinary bladder (4,25), intercorrelation of nuclear chromatin features in breast cancer (221, and the texture of the chromatin pattern of nuclei in endometrial hyperplasia and carcinoma (7).Apart from the presence of multilayers of cells, several factors complicate the quantitative analysis of tissue sections.

show abstract

Field test results using the BioPEPR cervical smear prescreening system

Cited by 16 publications

References 2 publications

Evaluation of contextual analysis for computer classification of cervical smears

Evaluation of contextual analysis for computer classification of cervical smears

Cytophotometric analysis of cervical intraepithelial neoplasia grade III, with and without synchronous invasive squamous cell carcinoma

Image processing for mitoses in sections of breast cancer: A feasibility study

Contact Info

Product

Resources

About